The `DeviceAdapter` provides an abstract interface to the accelerator
devices worklets and other algorithms run on. As such, the programmer has
less control about how the device launches each worklet. Each device
adapter has its own configuration parameters and other ways to attempt to
optimize how things are run, but these are always a universal set of
options that are applied to everything run on the device. There is no way
to specify launch parameters for a particular worklet.
To provide this information, VTK-m now supports `Hint`s to the device
adapter. The `DeviceAdapterAlgorithm::Schedule` method takes a templated
argument that is of the type `HintList`. This object contains a template
list of `Hint` types that provide suggestions on how to launch the parallel
execution. The device adapter will pick out hints that pertain to it and
adjust its launching accordingly.
These are called hints rather than, say, directives, because they don't
force the device adapter to do anything. The device adapter is free to
ignore any (and all) hints. The point is that the device adapter can take
into account the information to try to optimize for itself.
A provided hint can be tied to specific device adapters. In this way, an
worklet can further optimize itself. If multiple hints match a device
adapter, the last one in the list will be selected.
The `Worklet` base now has an internal type named `Hints` that points to a
`HintList` that is applied when the worklet is scheduled. Derived worklet
classes can provide hints by simply defining their own `Hints` type.
nvcc doesn't like the couple of places that we have Impl classes hiding behind a
private declaration. So for VTKM_CUDA only, ifdef the private declaration out
The number of bits in a `BitField` cannot be directly implied from the
size of the buffer (because the buffer gets padded to the nearest sized
word). Thus, the `BitField stored the number of bits in its own
internals.
Unfortunately, that caused issues when passing the `BitField` data
between it and an `ArrayHandleBitField`. If the `ArrayHandleBitField`
resized itself, the `BitField` would not see the new size because it
ignored the new buffer size.
To get around this problem, `BitField` now declares its own
`BufferMetaData` that stores the number of bits. Now, since the number
of bits is stored in the `Buffer` object, it is sufficient to just share
the `Buffer` to synchronize all of the state.
If you gave ReduceByKey a fancy output array that decorated another
array, you could get a runtime error for using an invalid array (if the
device adapter used the generic algorithm). The problem was that
ReduceByKey creates a temporary array, and that array was given the same
storage as the output array. That might not be valid for fancy arrays,
so instead use the default storage for the temporary array.
If you gave ScanInclusiveByKey a fancy output array that decorated
another array, you would get a runtime error for using an invalid array.
The problem was that ScanInclusiveByKey creates a temporary output array
and then copies the result to the actual output array. The problem was
that the temporary output array was given the same storage as the output
array, which won't work if the output array is fancy. Instead, make the
storage for the temporary array default.
The `ArrayHandleStreaming` class stems from an old research project
experimenting with bringing data from an `ArrayHandle` in parts and
overlapping device transfer and execution. It works, but only in very
limited contexts. Thus, it is not actually used today. Plus, the feature
requires global indexing to be permutated throughout the worklet
dispatching classes of VTK-m for no further reason.
Because it is not really used, there are other more promising approaches
on the horizon, and it makes further scheduling improvements difficult,
we are removing this functionality.
Marked the old versions of PrepareFor* that do not use tokens as
deprecated and moved all of the code to use the new versions that
require a token. This makes the scope of the execution object more
explicit so that it will be kept while in use and can potentially be
reclaimed afterward.
This behaves just like `ScanExclusive`, but rather than returning the
total sum, it is appended to the end of the output array.
This is in preparation for the CellSetExplicit refactoring described in
issue #408.
- Use AtomicInterface to implement device-specific atomic operations.
- Remove DeviceAdapterAtomicArrayImplementations.
- Extend supported atomic types to include unsigned 32/64-bit ints.
- Add a static_assert to check that AtomicArray type is supported.
- Add documentation for AtomicArrayExecutionObject, including a CAS
example.
- Add a `T Get(idx)` method to AtomicArrayExecutionObject that does
an atomic load, and update existing CAS usage to use this instead
of `Add(idx, 0)`.
BitFields are:
- Stored in memory using a contiguous buffer of bits.
- Accessible via portals, a la ArrayHandle.
- Portals operate on individual bits or words.
- Operations may be atomic for safe use from concurrent kernels.
The new BitFieldToUnorderedSet device algorithm produces an ArrayHandle
containing the indices of all set bits, in no particular order.
The new AtomicInterface classes provide an abstraction into bitwise
atomic operations across control and execution environments and are used
to implement the BitPortals.
Generalize DeviceAdapterAlgorithm::Transform to accept input array of different value and storage type.
Add doxygen documentation in DeviceAdapterAlgorithm.h
The implementation of the simplified version of
DeviceAdapterAlgorithmGeneral::Unique had two errors.
First, the implementation is such that it calls the more complex version
of Unique (which specifies a binary predicate to establish equality).
However, it was not calling the Unique method in the DerivedAlgorithm
like it should have been. Instead, it was calling its own Unique
algorithm, which might not be as efficient as the specialized Unique for
the device.
Second, it was using std::equal_to as its binary predicate. Using
functors from std can be dangerous because they are not marked with
VTKM_EXEC, so have the potential to not work in the execution
environment. Instead, use the readily available vtkm::Equal binary
predicate.
The implementation of ScanExclusiveByKey in
DeviceAdapterAlgorithmGeneral by shifting values in the input values
array and then calling ScanInclusiveByKey. However, the temporary
shifted values array was created using the key type instead of the
values type. This caused a compile error when the keys and values had
different types.
Often times we don't care about the output keys, and it's useful to
be able to pass an ArrayHandleDiscard into the algorithm to save
memory in these cases.
Sandia National Laboratories recently changed management from the
Sandia Corporation to the National Technology & Engineering Solutions
of Sandia, LLC (NTESS). The copyright statements need to be updated
accordingly.
By using the auto keyword and decltype we can reduce the number of
complex typedefs that exist when writing device adapter algorithms.
The goal being that it is easier for developers to see the actual
algorithms being implemented, by reducing the amount of template
'noise'.
Previously ArrayHandleReverse would only work if it was provided an explicit
users array to map too. But this doesn't need to be so, if a user wants to
start by constructing an ArrayHandleReverse we should allow that.
The side effect of this, is that some very tricky code in the DeviceAdapters
can be removed, that explicitly was added to allow output to ArrayHandleReverse
af61c7e6 Merge branch 'array_handle_reverse_write_2' of gitlab.kitware.com:ollielo/vtk-m into array_handle_reverse_write_2
869c9789 Merge branch 'array_handle_reverse_write_2' of gitlab.kitware.com:ollielo/vtk-m into array_handle_reverse_write_2
14592b82 Merge branch 'master' of gitlab.kitware.com:vtk/vtk-m into array_handle_reverse_write_2
1c50c86a remove commented out code to print values to stdout
80d4ee90 Enable output to ArrayHandleReverse
f64cb5ef Added test for using ArrayHandleReverse as output of ScanInclusiveByKey
87fe6768 thruw ErrorBadType, add PrepareForOutput from Pat
60542804 fixed a typo and added write support for ArrayHandleReverse
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !783
Redesigns the TBB and Serial backends and the vtkm::exec::Task concept so that
we can re-use the same launching logic for all Worklets, instead of generating
per worlet code. To keep the performance the same the TilingTask now is past
a range of indices to work on, rather than a single index.
Binary size reduction:
WorkletTests_SERIAL old - 19MB
WorkletTests_SERIAL new - 18MB
WorkletTests_TBB old - 39MB
WorkletTests_TBB new - 18MB
libvtkAcceleratorsVTKm old - 48MB
libvtkAcceleratorsVTKm new - 19MB